Advanced search  
Pages: [1] 2 3 ... 6   Go Down

Author Topic: Post-Loss Radio Signals  (Read 110086 times)

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Post-Loss Radio Signals
« on: September 23, 2011, 11:24:38 AM »

At long last our Catalog and Analysis of Post-Loss Radio Signals During the Search for Amelia Earhart in June 1937 is ready for prime time.  The amount of work that has gone into collecting, compiling, cataloging and analyzing these signals boggles the mind.  Over the past twenty years, dozens of TIGHAR researchers have contributed to this important aspect of our investigation of the Earhart disappearance but we want to especially thank LCDR Robert Brandenberg, USN (ret) whose tireless dedication and extraordinary expertise have made this catalog possible.

Radio signals, although invisible and fleeting, are just as real as bones and artifacts.  This is the first comprehensive compilation of what signals were reported and which of them are credible.  The importance of the story they tell is difficult to overstate.  This catalog is an essential tool in piecing together that story.  Take a look.
Logged

Ted G Campbell

  • TIGHAR member
  • *
  • Posts: 344
Re: Post-Loss Radio Signals
« Reply #1 on: September 23, 2011, 12:50:05 PM »

Ric,
Re the post loss signal catalog.

Have you and Bob B. discussed how it may have been possible for AE to pick up the KGMB broadcast, and apparently responded, but being unable to hear the Itasca at any point in the flight.

What radio would AE have used to hear KGMB?

Ted Campbell
Logged

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Re: Post-Loss Radio Signals
« Reply #2 on: September 23, 2011, 01:19:02 PM »

What radio would AE have used to hear KGMB?

Her Western Electric 20B receiver - the only receiver she had on the airplane.  We know the receiver was working because she heard the Morse code "A"s that Itasca sent on 7500 Khz. She heard that signal over her loop antenna, so we know the loop was working.  She didn;t hear anything else from Itasca because she switched back to the (missing) belly antenna.  To hear KGMB, all she had to do was realize that her belly antenna was gone and use the loop. Once she was on the ground she could get out of the airplane and see that the belly antenna was gone.
Logged

Ted G Campbell

  • TIGHAR member
  • *
  • Posts: 344
Re: Post-Loss Radio Signals
« Reply #3 on: September 23, 2011, 08:17:15 PM »

Ric,
So what you are sugesting is that AE's loop antenna was capable of receiving voice signals, this assumes that the KGMB was by voice, yet she didn't receive/reply to Itasca's voice transmissons.  Strange!

The question then becomes was the Itasca's radios operating correctly?

Ted Campbell
Logged

Jeff Scott

  • T2
  • **
  • Posts: 93
Re: Post-Loss Radio Signals
« Reply #4 on: September 23, 2011, 09:03:37 PM »

Ted,

I think the source of your confusion is Earhart was listening for the Itasca's transmissions on the belly antenna.  If it were missing or damaged in some way, this would explain why she apparently never heard any of the ship's voice transmissions.  She only heard the ship upon switching to the loop antenna while trying to take a bearing.  She then switched back to the belly antenna to for voice reception and apparently continued to be frustrated by not hearing anything.

If she truly did hear the KGMB broadcast, either she managed to repair the belly antenna or realized it was gone and relied on the loop antenna for receiving voice whereas she had not done so during flight.
It's not too late to be great.
 
Logged

Gary LaPook

  • T5
  • *****
  • Posts: 1624
Re: Post-Loss Radio Signals
« Reply #5 on: September 23, 2011, 11:02:50 PM »

Ric,
Re the post loss signal catalog.

Have you and Bob B. discussed how it may have been possible for AE to pick up the KGMB broadcast, and apparently responded, but being unable to hear the Itasca at any point in the flight.

What radio would AE have used to hear KGMB?

Ted Campbell
---------------------------------

Did Brandenburg do an analysis of the probability of Earhart on Gardner being able to hear the 1 kw signal from KGMB at the times that she would have needed to hear it in order to reply to it. He did such an analysis of its probability of being heard on the mainland to eliminate the possibility of hoaxters replying the the KGMB broadcasts.

gl
Logged

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Re: Post-Loss Radio Signals
« Reply #6 on: September 24, 2011, 06:37:09 AM »

So what you are sugesting is that AE's loop antenna was capable of receiving voice signals, this assumes that the KGMB was by voice, yet she didn't receive/reply to Itasca's voice transmissons.  Strange!

Not at all strange.  The loop is just an antenna.  It can receive voice just as well as code.  KGMB was (and still is) a commercial broadcast station.  Their broadcasts were in voice.  Earhart didn't hear Itasca's voice transmissions because she was listening for them on her missing belly antenna.

The question then becomes was the Itasca's radios operating correctly?

Itasca's transmitter seems to have been working correctly but there is considerable evidence that there were problems with the ship's receiver or receiving antenna system.  Itasca did not hear some of the post-loss signals that other stations (including Howland and Baker) did hear. 
Logged

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Re: Post-Loss Radio Signals
« Reply #7 on: September 24, 2011, 12:51:10 PM »

Did Brandenburg do an analysis of the probability of Earhart on Gardner being able to hear the 1 kw signal from KGMB at the times that she would have needed to hear it in order to reply to it.

Yes.  Bob reports:
The Signal to Noise Ration (SNR) was 56.6 dB/Hz.  The SNR required for 90% understandability of words and phrases is 45 dB/Hz, so the KGMB signal at Gardner was nearly 12 dB above the required level.  Since the SNR is well above the required level, we can safely say the probability of Earhart being able to hear KGMB was close to 100%.
Logged

Monty Fowler

  • T5
  • *****
  • Posts: 1078
  • "The real answer is always the right answer."
Re: Post-Loss Radio Signals
« Reply #8 on: September 24, 2011, 07:41:02 PM »

"The post-loss radio signals are the elephant in the room that can no longer be ignored."

*pictures Tom Crouch shaking his head in a dismissive way and sayiing something along the lines of, "Impossible!" in a dismissive and/or condescending tone, based solely on observations gleaned during multiple watchings of Finding Amelia*

That is NOT an elephant, Mr. Gillespie. That is, a, ummmm, largish gray thingy, you probably didn't see it because you weren't looking at the right spot at the right time of day, no, sunspots! Yeah, that's it, sunspots got in the way! Anyway, definitely not an elephant ... Nothing to see here folks, move along, please ...

Ex-TIGHAR member No. 2189 E C R SP, 1998-2016
 
Logged

Jeff Scott

  • T2
  • **
  • Posts: 93
Re: Post-Loss Radio Signals
« Reply #9 on: September 25, 2011, 11:54:01 AM »

This is indeed an excellent compilation of radio transmissions.  After reading through them all, I do have a question.

Several amateur reports from the west coast (usually claimed to be heard on 3105) are dismissed as "not credible" with one reason being that no one in the central Pacific who was much closer to the source and specifically listening on that frequency heard them.  Yet reports from people like Mabel Larremore in Texas, Nina Paxton in Kentucky, Mrs. Crabb in Toronto (July 5 & 6), Betty Klenck in Florida, and Thelma Lovelace in New Brunswick (while confirmed or presumed to be listening to harmonic frequencies of 3105) are considered credible even though no nearby Pacific stations heard anything.  Shouldn't these Pacific locations have heard something on 3105 around the same time as well?  I'd be tempted to demote at least some of these to "uncertain" since they are so isolated in time from signals received by other stations.

These amateur reports are among the strongest considering there were numerous locations in and around the Pacific that reported signals near the same time:

~1500Z July 4: Dana Randolph in Wyoming and Mrs. Crabb in Toronto simultaneously with Pan Am stations at Midway and Oahu

~0915Z July 5: Howard Coons in San Francisco within minutes of Navy stations at Howland, Baker, and San Francisco Division
It's not too late to be great.
 
Logged

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Re: Post-Loss Radio Signals
« Reply #10 on: September 25, 2011, 12:16:17 PM »

Several amateur reports from the west coast (usually claimed to be heard on 3105) are dismissed as "not credible" with one reason being that no one in the central Pacific who was much closer to the source and specifically listening on that frequency heard them.  Yet reports from people like Mabel Larremore in Texas, Nina Paxton in Kentucky, Mrs. Crabb in Toronto (July 5 & 6), Betty Klenck in Florida, and Thelma Lovelace in New Brunswick (while confirmed or presumed to be listening to harmonic frequencies of 3105) are considered credible even though no nearby Pacific stations heard anything.  Shouldn't these Pacific locations have heard something on 3105 around the same time as well?

Receptions of a signal at the same time by multiple stations is certainly a big credibility plus but it rarely happened, even among stations in the Central Pacific. In truth, Itasca and Coast Guard Hawaiian Section (COMHAWSEC) were the only stations in the Central Pacific maintaining a constant watch on 3105.  Pan Am, Navy Wailupe, and Navy Tutuila listened from time to time but had other business to tend to.  Itasca missed messages it should have heard (apparently had receiver or antenna problems) and COMHAWSEC was a long way from Gardner.  When Itasca heard unintelligible voice, COMHAWSEC either heard nothing or just a carrier wave.   

In other words, lack of simultaneous reception by other stations is not a enough to throw a signal into the "uncertain" category. The amateurs we consider to be credible have  both quantitative and qualitative factors in their favor.  The frequency and probability are in the credible range and the content of the reported messages are consistent with known constraints.
Logged

Jeff Scott

  • T2
  • **
  • Posts: 93
Re: Post-Loss Radio Signals
« Reply #11 on: September 25, 2011, 06:23:16 PM »

Your points are well taken, which leads to the question of whether other stations hearing anything around the same time is even a valid criterion in the first place.  My fear is that the overall work compiled by you and Mr. Brandenburg might be discounted by naysayers because of the seeming double standard in how credibility is determined.  I don't believe you are doing this at all, but I could see others who are promoting their own theories making the argument that reports supporting the Gardner hypothesis are deemed "credible" while any that contradict it are deemed "not credible."  On the contrary, I believe you've made a convincing case that the aggregate of these reports supports the case of a landing at Gardner regardless of whether individual examples like Betty Klenck or Mabel Larremore are labeled as "credible" or "uncertain."

Here are some ideas that might strengthen the overall quality of the catalog:

1) Perhaps the "Qual Factors" could be clearly broken into pros and cons listing factors in favor of the transmission being a legitimate Earhart message and those opposed.  Whether it be the "LA hoaxers" and Charles Miguel or Betty, Mabel, Nina, etc., lack of reception by other stations around the same time would be a con.  The former group has additional factors against it while the latter has other factors in favor which help make a more balanced judgement of credibility.

2) I think I struggle with the concept of "credible" because it is so binary.  Something can be barely credible or extremely credible yet it is "graded" equally in the catalog.  Another idea to consider is changing from "credibility" to "level of confidence" that the signal was from the Electra.  This could include 4 levels instead of the 3 used now:

- High: Probability of reception is good (maybe >0.2) and there are several qualitative factors that would be considered "pros" far outweighing the "cons" against it.  These would be the stronger of the reports listed as "credible."

- Medium: Probability of reception is plausible (maybe 0.001 to 0.2) and the "pros" seem to outweigh the "cons."  These would be the weaker of the reports listed as "credible."

- Low: Probability of reception is low and/or pros and cons are roughly equal.  These are the reports called "uncertain."

- None: Probability is essentially 0 and/or cons far outweigh the pros.  These are report now called "not credible."
It's not too late to be great.
 
Logged

Bob Brandenburg

  • TIGHAR member
  • *
  • Posts: 33
Re: Post-Loss Radio Signals
« Reply #12 on: September 26, 2011, 02:21:38 PM »

Jeff,

The question of whether other stations heard something around the time of a signal is relevant in some  circumstances, but not all.  For example, a report by a source in the continental U.S. claiming to have heard Earhart on 3105 would be suspect if none of the numerous airport stations listening 24/7 on that frequency heard it.  On the west coast, COMFRANDIV had a dedicated 24/7 watch on 3105 and 6210, using high-gain antennas aimed at Hawaii -- and by extension the central Pacific -- and it would be extremely unlikely for some one on the coast to hear a signal on either frequency without COMFRANDIV hearing it.

The notion of degrees of credibility related to reception probability suggests confusion about the meaning of probability.  The probability in the catalog reflects the factors affecting signal-to-noise ratio (SNR) at the receiver at the time.  For example, a probability of 0.002 means that, given the ionospheric conditions at the time, random variability in propagation factors would cause the SNR to rise above the reception threshold on 2 occasions out of 1,000.   But whether the event in question was believably one of the 2 times depends on the reported signal content, reporting party credibility, etc.

If a signal was heard, it was heard regardless of the propagation probability.  A signal on a low probability  path can fade in briefly, or for an extended time, then disappear.  A signal on a high probability path can fade out.   International shortwave broadcasters minimize fading dropouts of their signals by transmitting  at power levels on the order of  tens of thousands of watts to overcome fading.   Earhart's low-power signals were at the mercy of the propagation conditions.

The idea of degrees of credibility is superficially attractive, but unworkable.  Ric and I went through this some time ago.  Credibility, like pregnancy, is inherently binary.  Either we believe something or we don't.   The credibility criteria in the catalog are essentially subjective, but when applied they yield a binary result.   Assigning intermediate degrees of credibility, based on whatever criteria, opens the door to endless quibbling about the appropriate degree for a given case, and what difference it makes. 

Bob
Logged

Ric Gillespie

  • Executive Director
  • Administrator
  • *
  • Posts: 6098
  • "Do not try. Do or do not. There is no try" Yoda
Re: Post-Loss Radio Signals
« Reply #13 on: September 26, 2011, 02:42:14 PM »

My fear is that the overall work compiled by you and Mr. Brandenburg might be discounted by naysayers because of the seeming double standard in how credibility is determined.

Naysaying is a mind set.  I've seen it at work among our own researchers and have come to understand that it has little to do with facts or reason. I've stopped worrying about naysayers.

  I don't believe you are doing this at all, but I could see others who are promoting their own theories making the argument that reports supporting the Gardner hypothesis are deemed "credible" while any that contradict it are deemed "not credible."  On the contrary, I believe you've made a convincing case that the aggregate of these reports supports the case of a landing at Gardner regardless of whether individual examples like Betty Klenck or Mabel Larremore are labeled as "credible" or "uncertain."

"Uncertain" is not a lesser grade of "credible."  When we say "uncertain" we're saying that there is not enough information to make a judgement about credibility.

You're right, it's the aggregate of all the findings that makes the case.  Our analysis tests the hypothesis that the signals came from the Earhart plane parked on the reef at Gardner.  In no case is the credibility of a signal influenced by whether or not it supports that hypothesis. It happens that all of the credible signals could have been sent from the reef at Gardner.  The hypothesis is therefore supported. Of course, some other unstated and untested hypothesis might also be supported but would have to pass the same rigorous tests.

1) Perhaps the "Qual Factors" could be clearly broken into pros and cons listing factors in favor of the transmission being a legitimate Earhart message and those opposed.  Whether it be the "LA hoaxers" and Charles Miguel or Betty, Mabel, Nina, etc., lack of reception by other stations around the same time would be a con.  The former group has additional factors against it while the latter has other factors in favor which help make a more balanced judgement of credibility.

Right now we discuss the "Qual Factors" for each signal in a short narrative paragraph. 
We could, I suppose, replace the paragraph with a checklist of pros and cons with each factor being assigned a value.  Some factors would be "gatekeepers" that are by nature binary.

Frequency?  The message must have been heard on 3105, 6210 or a harmonic of those frequencies.  If it wasn't it's an automatic "Not Credible."
Well-sent code? Automatic "Not Credible."
Description of a floating airplane? Automatic "Not Credible."
And so forth.

Other factors are trickier and would require that we assign a range of values that would be, by definition, arbitrary.  For example,
- What is simultaneous reception by another station worth?  One positive point?  Five positive points?
- What is it worth if dashes are heard immediately following a request for dashes?   One positive point?  Five positive points?
- What if an otherwise credible amateur later changes her story, as Paxton did, to include information that is not credible? One negative point?  Five negative points?

The result would be to make each catalog entry interminably long and create the impression that we're trying to quantify factors that are unavoidably qualitative.  The catalog cannot hope to present the entire case for or against each signal.  Naysayers will undoubtedly dismiss our conclusions about the post-loss radio signals with generalities just as they do with our archival evidence and artifacts.  We can defend our evaluation of any signal with reams of specifics.  I think it's best to leave our "Qual Factors" just as they are.

2) I think I struggle with the concept of "credible" because it is so binary.  Something can be barely credible or extremely credible yet it is "graded" equally in the catalog.  Another idea to consider is changing from "credibility" to "level of confidence" that the signal was from the Electra.

I'm not sure you understand what we mean by "credible." It does not mean “proven to be authentic.”  The only way to prove that a signal was an authentic communication from Amelia Earhart would be to ask her if she sent it and, given our aversion to psychics, we haven't found a way to do that.  "Credible" means we haven't found a reason that it couldn't be a genuine signal from the Electra. There are no gradations.  Either the message is disqualified or it's not. 

« Last Edit: September 26, 2011, 02:45:34 PM by Ric Gillespie »
Logged

Monty Fowler

  • T5
  • *****
  • Posts: 1078
  • "The real answer is always the right answer."
Re: Post-Loss Radio Signals
« Reply #14 on: September 26, 2011, 08:27:09 PM »

Sooooo, just out of curiosity, what, if anything, does Tom Crouch have to say about these signals?

LTM,
Monty Fowler
TIGHAR No. 2189CE
Ex-TIGHAR member No. 2189 E C R SP, 1998-2016
 
Logged
Pages: [1] 2 3 ... 6   Go Up
 

Copyright 2024 by TIGHAR, a non-profit foundation. No portion of the TIGHAR Website may be reproduced by xerographic, photographic, digital or any other means for any purpose. No portion of the TIGHAR Website may be stored in a retrieval system, copied, transmitted or transferred in any form or by any means, whether electronic, mechanical, digital, photographic, magnetic or otherwise, for any purpose without the express, written permission of TIGHAR. All rights reserved.

Contact us at: info@tighar.org • Phone: 610-467-1937 • Membership formwebmaster@tighar.org

Powered by MySQL SMF 2.0.18 | SMF © 2021, Simple Machines Powered by PHP